200 Hose-in-Hose Flowers [ch. 



appear, but owing to the great fluctuation in the degree of 

 doubleness exact counts are impossible. 



In Carnation, where the doubhng is of the ordinary 

 kind, due to petalody combined with sub-division of parts, 

 there is some reason for suspecting doubleness to be a 

 dominant. Everyone who has grown Carnations from seed 

 is aware that a proportion, often considerable, of the seed- 

 lings come single. It is most improbable that any large 

 number of these can owe their singleness to cross-fertilisa- 

 tion with single plants, for breeders would not keep such 

 plants wittingly. Till critical experiments are made, how- 

 ever, the point cannot be regarded as certain beyond 

 question. 



In Poppies, on the contrary, the dominance of the single 

 type appears to be quite complete. In the annual Larkspur 

 {^Delphinium consolida) also, from the fact that pink doubles 

 exposed to the pollen of blue singles produce among their 

 seedlings some blue singles'^, there is no doubt that the 

 single type is dominant. Instances of this sort could 

 probably be multiplied without difficulty, and from what we 

 know of plant-breeding in general, there is no practical 

 doubt that doubling of the ordinary type is usually recessive 

 to singleness. 



The Hose-in-Hose or Cafycanlhemott^s Campanula and 



Mimulus. 



The inheritance of this well-known variation or mon- 

 strosity was studied by Correns (76). The calyx, as 

 described above, is petalold in many degrees, varying from 

 a condition in which the sepals are still partially green up 

 to the full development of a second corolla. In Correns' 

 experience this variation In the calyx is accompanied by a 

 marked diminution in the fertility of the female organs. 

 This in Campanula persicifolia amounted to total sterility, 

 though in C, media a few seeds were set. 



The abnormal character proved to be a partial and 

 somewhat irregular dominant, considerable fluctuation occur- 

 ring on the individual plants. Owing to complete failure of 

 F^ to set seed F^ could not be raised, but when the normal 

 type was fertilised with pollen from F^ a mixture of the two 



* I have observed this in my own garden. 



